hanslovsky

Assume you have an indexed Spark RDD of N images and you would like to sparsely select pairs, i.e. the number of selected pairs is much smaller than all possible pairs N^2. One way to do it, is to create a cartesian of the RDD with itself and then filter the pairs by some criterion, e.g. limit the difference between pairwise indices:

selection = images
    .cartesian( images )
    .filter( /* some criterion */ )
    ;

However, this needs to create an intermediate RDD with N^2 elements and on the way their requires a lot of cross-talk between

hanslovsky

hanslovsky

hanslovsky

hanslovsky

import java.util.stream.Collectors;
import java.util.stream.IntStream;

import net.imglib2.Cursor;
import net.imglib2.Interval;
import net.imglib2.Point;
import net.imglib2.RandomAccess;
import net.imglib2.RandomAccessible;
import net.imglib2.RandomAccessibleInterval;
import net.imglib2.algorithm.neighborhood.DiamondShape;

hanslovsky

hanslovsky

hanslovsky

I assume now that these lines in jnius/jnius_proxy.pxi and these lines cause the issue:

cdef jobject py_invoke0(JNIEnv *j_env, jobject j_this, jobject j_proxy, jobject
        j_method, jobjectArray args) except * with gil:

    from .reflect import get_signature, Method
    cdef jfieldID ptrField
    cdef jlong jptr

hanslovsky

Install imglyb Through Conda

If you are running Linux or OSX you can install imglyb through conda:

# install conda through your package manager or from
# https://conda.io/docs/download.html
conda create -n imglyb -c hanslovsky imglib2-imglyb imglyb-examples python=3.6
source activate imglyb

# run examples (Linux)
python -m imglyb-examples.butterfly

hanslovsky

Not sure how this works but this might point you in the right direction @kephale